Vienna Symphonic Library Forum
Forum Statistics

194,476 users have contributed to 42,922 threads and 257,973 posts.

In the past 24 hours, we have 3 new thread(s), 13 new post(s) and 79 new user(s).

  • Sympathetic Vibrations Idea

    Herb and all at VSL,

    I may have come up with an idea of how to tackle the problem of sympathetic vibrations either between ensembles, or created when 16 solo violin sample instruments play together as an ensemble. The problem with using say, 16 different violin instruments samples together as an ensemble in traditional sample rigs is that no sympathetic vibrations occur between the instruments creating cross related harmonics etc... For those not familiar with the concept you should be able to read more by doing a search on Google.

    Anyhow, here's the solution, assuming you have 16(or a set of many) seperate instrument samples intended to be used in soli unison (or really any which way):

    1. Put local mics (short range to avoid picking up neighboring instruments but mostly nearby wall reflections) on all the seperate instruments and set them up in a silent room.

    2. Either use a shot gun pistol or wave sweep from two locations:
    a) first location would be 25% inside center from the left side of the group
    b) 2nd location would be 75% inside center or 25% from right side of group

    Here's the part that is going to do it:
    3. Use the captured recording of the instruments vibrations from the sound sweep or pistol as a reverb map in AltiVerb from AudioEase. Sure it takes a lot of processing power, but this should work.

    4. Add this AltiVerb vibration to the appropriate ensemble and whamo, you should have the appropriate resonances inherent.

    This method could also be used for the whole orchestra as well as for neighboring sections such as the brass instruments vibrating and reacting to the same gunshot heard within the violin section. In fact you could have a WHOLE ORCHESTRA vibration capture for each section of the orchestra. Meaning, you put local mics on ALL the instruments that could possibly be on stage, including timpani etc... and you mix them accordingly so that they are proportional to the location of either a conductor or audience member, THEN, you fire the gunshot from within the:
    violins
    violas
    celli
    basses
    horns
    trombones
    flutes
    percussion
    etc...

    You'd then slap the apprpriate WHOLE ORCHESTRA vibration as a patch in AudioEase's AltiVerb, and viola, you have the apprpriate resonances to each frequency possible for each section playing.

    Of course you'd need to setup like 10 to 20 AltiVerbs, but this is not a problem for offline bouncing within Logic for instance.

    I just thought I'd share this in case someone wants to give it a shot. Just remember me when you mention where the idea came from (that is if it works, but I suspect it will).

    This approach is similar to Physical Modeling except that the resonance of the instruments is sampled as opposed to theoretically and mathematically constructed.

    It should work!

    Perhaps you guys can do some tests, and maybe venture into this area, staying ahead of others who may try...

    Yours,
    Evan Evans
    evanevans@evanevans.org

  • This seems like a lot to go through in proportion to how much better it'll make things sound.

    Is this resonance effect really that noticable? I can understand something like snare drum sympathetic vibration, but how much does this occur in other instruments? Can you really tell the difference between a violinist playing in a room full of violins, and the violinist playing alone in the same room?

    Things like recording brass with mutes are way more important than this.
    It is worth it though as a very last addition to the library, if there is going to be a "last addition".

    Anthony Lombardi
    www.mp3.com/alombardi

  • What do you think of having ambiance noises from real orchestral hall that we could add to our vienna lib to make things dirtier, more alive.. (creaking chair, noozing audience...)?
    You know the kind of non-silence you can hear on real records..

    I am sure it works,
    Don't we sometimes add analog noise to warm up some digital tracks?

    Laurent

  • What do you think of having ambiance noises from real orchestral hall that we could add to our vienna lib to make things dirtier, more alive.. (creaking chair, noozing audience...)?
    You know the kind of non-silence you can hear on real records..

    I am sure it works,
    Don't we sometimes add analog noise to warm up some digital tracks?

    Laurent

  • There may be a flaw in your theory, Evan

    the sound of an instrument next to another instrument affects the way the sound is actually made, with sympathetic vibrations etc. The initial samples for each instrument are not affected in this way by Altiverb - only with a reaction to the wave-sweep or gun, and not to the individual notes. Violins, for instance, resonate differently with each note (the belly resonates to a D, and the back to E [or vice-versa] for a Strad-based model). The AltiVerb can only give an example of the ambience at that point in the room...

    still, it might be worth a go - you gonna fund it? [[;)]]

    Peter

  • To all -

    although I don't want to make too much noise (pun intended ;-]) about all this yet, I think I'm allowed to tell you that we work already on a solution similar to these suggested concepts since almost a year now.

    Watch this space for news later this year ... interesting times are ahead, I asure you ;-]

    /Dietz - Vienna Symphonic Library

    /Dietz - Vienna Symphonic Library
  • Right now I'd just be happy with some sort of automated sample randomizer...

    I've spent the last 2 weeks on a piece sampling myself playing sustained drones on a kazoo, looped them, and then mapped these across the entire stereo field (near, far, and evetything in between). However, to simulate a true chorus of instruments (I'm using 20 in total at this point), I had to route the samples through pitch, EQ, phase and amplitude envelope effect plugins--using unique settings for each sample--before mixdown. The result is a surprisingly realistic ensemble texture that is consistent in its timbral inconsistency.

    So.

    It seems to me that there should be some method of automating this "humanizing" effect--routing a given sample through a preset "macro" of audio manipulations which would randomize itself within user-defined limits to achieve believable randomness (some sequencers already do this with MIDI data). This could then be applied to any collection of samples without the need to do it manually, sample by sample.

    Of course, reverberative resonance and sympathetic vibrations would be the final great frontiers in achieving ensemble-quality textures from solo instruments..!

    My 2 cents [*-)]

  • last edited
    last edited

    @topcomposer said:

    There may be a flaw in your theory, Evan
    the sound of an instrument next to another instrument affects the way the sound is actually made, with sympathetic vibrations etc. The initial samples for each instrument are not affected in this way by Altiverb - only with a reaction to the wave-sweep or gun, and not to the individual notes. Violins, for instance, resonate differently with each note (the belly resonates to a D, and the back to E [or vice-versa] for a Strad-based model). The AltiVerb can only give an example of the ambience at that point in the room...


    Your welcome to contradict me. I enjoy conjecture and debate.

    Actually when you run a white noise or better yet, the filter sweep through something that will give off re-verberations (or vibrations), you get the frequency response PER focus frequency. So really, a filter sweep gives you the reactions to each frequency.

    With AlitVerb if you play some low ContraBasses and you have a big sampled hall setup, you will get a different reverberation response than if you play shrieking violins. I am not talking pitch wise, but actually the file that AltiVerb uses based on a filter sweep live in the hall has the information in it to show how a high frequency sound would reverberate compared to a low frequency sound, and everything elese swept inbetween.

    So It really is the answer to capturing the full frequency response to every ffrequency you can throw at it. It's kind of like a dynamic frequency response, but the dynamic part refers to the shift in frequency rather than volume (power).

    A revision to my theory may be to use local vibration pickups on an unobtrusive part of the body of the instrument(s) to achieve maaximum insrtument to non-instrument rejection, so that the result is purely the vibration of the instruments without any wall reflections or other ambiences (other than what would effect the instrument ie: some wall reflections probably can be felt in the instrument's vibrations).

    I'm not sure if I fully addressed your question, but I believe there was something amiss in your assessment of this technology, mainly regarding how the captured file would react to a D3 differently than say A#6, and it would.

    Evan

  • last edited
    last edited

    @Another User said:

    Things like recording brass with mutes are way more important than this.
    It is worth it though as a very last addition to the library, if there is going to be a "last addition".


    I absolutely 100% agree. I prefer we get VSL completely exhaustive of 80 to 90% of what an orchestra can do, then we can focus on this whole other venture really, of having individual players to make your own orchestra. This is more experimental, but is the VSL currently really not that experimental? Or GOS or Dan Dean or any other HUGE library? I submit that VSL is already on an experimental mission, and that they will need to think up new refreshing ways to continue to voyage to where no man has gone before, or else they won't have anyhting fun to do in a decade from now. [:'(]

    Evan

  • evan

    You may, indeed be right! I haven't used AltiVerb yet, so I may not have my head around the maths properly... still, I'm quite excited by the pre-news from Dietz - I think we're all going to spend SOOOO much money on these guys in the coming months/years...!

    peter

  • In addition to the physical resonance of the instrument bodies, there is another dynamic property of instrument groups: Constructive and Destructive Interference of the instruments waveforms. Some of this may be realized in the summing of waveforms in the digital domain.